Generic Instrument
We have added a way to add "any spectrum" to LBL without having a pre-defined instrument.
Note that once you get this to work you can submit your wrap file and we can add it as a full instrument fairly easily.
Here are the following steps to setup a generic instrument.
Please do not try this unless you have successfully run a support instrument through LBL.
Step 1: Run lbl setup
Run LBL setup with the following command:
lbl_setup {my_wrap_file}.py
where {my_wrap_file} is the name if the wrap file you want to create
When asked for an instrument type "Generic".
This should create a wrap file similar to those for normal instruments but with many "other settings" that are set by default for our supported instruments.
Step 2: Modify the wrap file
The wrap file starts off very similar to the supported instruments (we recommend you run a normal instrument once before trying to add a new Generic instrument).
You'll see under the normal settings a large section:
# -------------------------------------------------------------------------
# Other settings
# -------------------------------------------------------------------------
These are required to run a generic instrument.
A detailed description of all wrap parameters is here.
Looking at the parameters for supported instruments may help in setting these values.
In the github directory look for lbl/instrument directory and find a method in the classes called "param_override":
# =============================================================================
# Define Espresso class
# =============================================================================
class Espresso(Instrument):
def __init__(self, params: base_classes.ParamDict):
# call to super function
super().__init__('ESPRESSO')
# extra parameters (specific to instrument)
self.default_template_name = 'Template_{0}_ESPRESSO.fits'
# define wave limits in nm
self.wavemin = 377.189
self.wavemax = 790.788
# set parameters for instrument
self.params = params
# override params
self.param_override()
# -------------------------------------------------------------------------
# INSTRUMENT SPECIFIC PARAMETERS
# -------------------------------------------------------------------------
def param_override(self):
"""
Parameter override for SPIRou parameters
(update default params)
:return: None - updates self.params
"""
# set function name
func_name = __NAME__ + '.Espresso.override()'
# set parameters to update
self.param_set('INSTRUMENT', 'ESPRESSO', source=func_name)
# add instrument earth location
# (for use in astropy.coordinates.EarthLocation)
self.param_set('EARTH_LOCATION', 'Paranal')
# define the default science input files
self.param_set('INPUT_FILE', '*.fits', source=func_name)
# The input science data are blaze corrected
self.param_set('BLAZE_CORRECTED', True, source=func_name)
# define the mask table format
self.param_set('REF_TABLE_FMT', 'csv', source=func_name)
# define the mask type
self.param_set('SCIENCE_MASK_TYPE', 'full', source=func_name)
self.param_set('FP_MASK_TYPE', 'neg', source=func_name)
self.param_set('LFC_MASK_TYPE', 'neg', source=func_name)
# define the default mask url and filename
self.param_set('DEFAULT_MASK_FILE', source=func_name,
value=None)
...
This will tell you what values we set for this instrument and data source combination.
Step 3: Creating data
Spectroscopic data comes in many forms. For supported instruments we support the output of those pipelines.
However for the Generic instrument mode we ask you to "convert" your data into a standard format that can be understood by LBL. We document this below.
You should create one fits file per observation.
Each fits file should have the following extensions
Primary extension: No data but a header (this can be the header from your input data) but must contain the values you set for
KW_MID_EXP_TIME, KW_MJDATE, KW_BERV, KW_BJD, KW_EXPTIME, KW_DATE, KW_OBJNAME, KW_AIRMASS, KW_SNR, KW_EXT_SNR
These are required to run a generic instrument.
FLUX: An Image HDU of shape x=number of pixels y=number of orders
WAVELENGTH: A wavelength grid matching the shape of FLUX
BLAZE: A blaze grid matching the shape of WAVELENGTH - this can be left out if setting BLAZE_CORRECTED = True in the wrap file
If you don't know keys such as "SNR" or "EXT_SNR" or "KW_AIRMASS" just set these to a sensible value.
An example of how to save a file like this in python is provided below:
from astropy.io import fits
import numpy as np
# Create an empty primary HDU with a header
primary_hdu = fits.PrimaryHDU()
primary_hdu.header['COMMENT'] = "Primary HDU is empty but contains a header."
# Create example 2D numpy arrays
flux_data = np.random.random((48, 2024)) # Example 48x2024 flux image
wavelength_data = np.random.random((48, 2024)) # Example 48x2024 wavelength image
# Create image HDUs for FLUX and WAVELENGTH
flux_hdu = fits.ImageHDU(data=flux_data, name='FLUX')
wavelength_hdu = fits.ImageHDU(data=wavelength_data, name='WAVELENGTH')
# Create an HDU list and write to a file
hdul = fits.HDUList([primary_hdu, flux_hdu, wavelength_hdu])
hdul.writeto('example.fits', overwrite=True)
print("FITS file 'example.fits' has been created successfully.")
These are required to run a generic instrument.
Step 4
Run your wrap file as normal:
python {my_wrap_file}.py
Supporting your instrument
From this wrap file + a code that converts your input data into our "generic format" we can add your instrument to the list of supported instruments very easily.
We do request a small set of ~20 files science files that can be used for demo and testing purpose (so any future changes are also supported).
Please email: neil.cook@umontreal.ca and etienne.artigau@umontreal.ca for more information or to request adding your instrument.